Citrus green fruit detection via improved feature network extraction

Jianqiang Lu; Ruifan Yang; Chaoran Yu; Jiahan Lin; Wadi Chen; Haiwei Wu; Xin Chen; Yubin Lan; Weixing Wang

doi:10.3389/fpls.2022.946154

Citrus green fruit detection via improved feature network extraction

Front Plant Sci. 2022 Nov 30:13:946154. doi: 10.3389/fpls.2022.946154. eCollection 2022.

Authors

Jianqiang Lu^{1

2

3}, Ruifan Yang^{1

3}, Chaoran Yu^{4

5}, Jiahan Lin^{1

3}, Wadi Chen^{1

3}, Haiwei Wu^{1

3}, Xin Chen^{1

3}, Yubin Lan^{1

2

3}, Weixing Wang^{1

6}

Affiliations

¹ College of Electronic Engineering (College of Artificial Intelligence), South China Agricultural University, Guangzhou, China.
² Guangdong Laboratory for Lingnan Modern Agriculture, Guangzhou, China.
³ National Center for International Collaboration Research on Precision Agricultural Aviation Pesticide Spraying Technology, Guangzhou, China.
⁴ Vegetable Research Institute, Guangdong Academy of Agricultural Sciences, Guangzhou, China.
⁵ Guangdong Key Laboratory for New Technology Research of Vegetables, Guangzhou, China.
⁶ Guangdong Provincial Agricultural Information Monitoring Engineering Technology Research Center, Guangzhou, China.

Abstract

Introduction: It is crucial to accurately determine the green fruit stage of citrus and formulate detailed fruit conservation and flower thinning plans to increase the yield of citrus. However, the color of citrus green fruits is similar to the background, which results in poor segmentation accuracy. At present, when deep learning and other technologies are applied in agriculture for crop yield estimation and picking tasks, the accuracy of recognition reaches 88%, and the area enclosed by the PR curve and the coordinate axis reaches 0.95, which basically meets the application requirements.To solve these problems, this study proposes a citrus green fruit detection method that is based on improved Mask-RCNN (Mask-Region Convolutional Neural Network) feature network extraction.

Methods: First, the backbone networks are able to integrate low, medium and high level features and then perform end-to-end classification. They have excellent feature extraction capability for image classification tasks. Deep and shallow feature fusion is used to fuse the ResNet(Residual network) in the Mask-RCNN network. This strategy involves assembling multiple identical backbones using composite connections between adjacent backbones to form a more powerful backbone. This is helpful for increasing the amount of feature information that is extracted at each stage in the backbone network. Second, in neural networks, the feature map contains the feature information of the image, and the number of channels is positively related to the number of feature maps. The more channels, the more convolutional layers are needed, and the more computation is required, so a combined connection block is introduced to reduce the number of channels and improve the model accuracy. To test the method, a visual image dataset of citrus green fruits is collected and established through multisource channels such as handheld camera shooting and cloud platform acquisition. The performance of the improved citrus green fruit detection technology is compared with those of other detection methods on our dataset.

Results: The results show that compared with Mask-RCNN model, the average detection accuracy of the improved Mask-RCNN model is 95.36%, increased by 1.42%, and the area surrounded by precision-recall curve and coordinate axis is 0.9673, increased by 0.3%.

Discussion: This research is meaningful for reducing the effect of the image background on the detection accuracy and can provide a constructive reference for the intelligent production of citrus.

Keywords: CB-Net; Mask-RCNN; deep learning; feature fusion; instance segmentation.